Platform vs Custom: How DMEs Should Actually Decide
By Irfan Mirza · May 2026 · 7 min read
Every DME owner I talk to in 2026 is being pitched the same idea: "Our AI platform will automate your intake, prior auth, resupply, and billing — let us migrate you off your current system." The platforms making that pitch are well-funded (GenHealth, Parachute Health, Tomorrow Health, and a half-dozen more), well-marketed, and increasingly visible at trade shows.
The pitch isn't wrong. Automation is the right answer. The platform-shaped path to automation, though, isn't the only path — and for most small-to-mid DMEs, it's not the cheapest or fastest one. Here's how to actually decide between buying a platform and building a custom automation layer on top of what you already run.
The honest comparison
Buy a platform when
- You're doing $5M+/month in billables and the cost of a platform migration amortizes inside 18 months.
- Your current system of record is genuinely failing — not just leaking — and a parallel platform investigation has already started.
- You have a CFO or COO who can quarterback a 6–12 month migration and absorb the cash-flow disruption.
- You're VC-backed, PE-backed, or actively scaling — the platform's enterprise pricing matches the capital structure you're operating inside.
- You're consolidating multiple DMEs onto a single platform and the migration is happening anyway.
Build a custom layer when
- You're doing $500K–$5M/month and Brightree / WellSky / NikoHealth / Bonafide are working as a system of record.
- Your gaps are workflow-shaped, not data-shaped — intake re-keying, auth tracking, resupply outreach, denial prevention.
- You want capex, not opex — own the automation, no enterprise renewal.
- You can't afford a 6-month implementation gap during which staff is partially trained and operations is partially broken.
- Your product mix is varied enough (CPAP + oxygen + enteral + urological, each with different LCDs) that a generic platform will force workflow compromises.
The five questions that decide it
Before you take a platform demo, answer these:
1. What is your current platform actually failing at?
If your Brightree or WellSky instance is recording the right data but the workflow on top is broken, you don't need a different platform — you need a different workflow layer. A new platform won't fix workflow gaps; it will just put them in a different UI.
2. What does a platform switch actually cost you?
License is the small number. Implementation, training, data migration, dual-running, productivity loss, payer re-credentialing on the new platform, staff turnover during the chaos — these are the real costs. A typical DME platform migration in our experience runs $80K–$400K all-in for a small/mid supplier, plus a 4–10 month productivity dip.
3. What does a custom layer actually cost you?
The four engagements that make up most DME workflow automation — intake validation, auth tracking, resupply outreach, billing pre-flight — totaled around $7,500–$16,000 (published pricing here) when bundled. Add 18% of that per year for ongoing care. Implementation is 12–16 weeks delivered in sequence. No platform migration, no productivity dip.
4. Who owns the automation when the contract ends?
Platform answer: nobody. The automation runs on the platform's infrastructure; if you leave, the automation leaves. Custom answer: you do. The automation lives on your infrastructure, integrates into your platform, and outlives any vendor relationship.
5. Who knows your specific workflow?
Platform answer: a sales engineer and an implementation consultant, both of whom are working on six other DMEs this quarter. Custom answer: someone who actually runs one — or doesn't, depending on who you hire. If you go custom, hire an operator-builder. If you can't find one, the platform option starts looking better.
The honest middle ground
You can do both. Some DMEs adopt a workflow platform and build custom layers around it. Others run their existing system of record (Brightree, WellSky) with a custom automation layer for years, then revisit the platform question as they scale past $5M/month.
The mistake is not the decision — it's not having one. Most DMEs we talk to are losing 15–25 hours per location per week to manual workflow, plus 25–35% in preventable denials, with no decision either way. The first move is to quantify the leakage. From there, the buy-vs-build question answers itself.
What to do next
If you want a fast, no-pitch quantification of your specific leakage: the free DME leakage audit is 30 minutes, run by an operator, and gives you a written summary the next day. It will tell you whether a platform investigation makes sense for you, whether a custom layer makes sense, or whether the right move is to fix one thing manually and revisit in six months.
Disclosure: SynergyIQ delivers the four productized custom engagements described above. We do not sell or resell any DME platform. The "buy a platform" path is not something we benefit from — and is sometimes the right answer for the DME we're talking to.